388 MR. J. J. MANLEY: OBSERVATIONS ON THE 



immediately before or after the completion of a comparison. The usual custom of 

 observing the pointer through a telescope was also followed. The limits of the 

 pointer's excursions to the right and left of the resting-point (R.P.) were estimated 

 in tenths of a scale division. 



The standardizations were effected by means of a Bunge short-beam balance, and, 

 in order to obtain a check upon the indicated values, it was decided to re-compare 

 several of the weights with the aid of an Oertling balance having a long beam. It 

 was then discovered that the two balances gave slightly, but distinctly, different 

 values for the several weights ; it was likewise found that values differing from each 

 other to a very small extent were generally indicated in each succeeding comparison 

 with either balance. 



On referring to numerous papers dealing with atomic weight determinations, and 

 other work involving very accurate weighing, it was found that many observers had 

 experienced similar difficulties. 



MILLER, in his " Determination of the Standard Pound,"* alludes to certain 

 progressive changes in the resting-point of his balances, and suggests causes to which 

 they may be due. 



During his observations for the determination of the " Mean Density of the Earth," 

 PoYNTfNGt was troubled by the shifting to and fro of the resting-point. THORPE, in 

 his paper on the atomic weight of titanium,{ draws particular attention to the 

 inexplicable behaviour of his Oertling short-beam balance. When determining the 

 atomic weight of boron, ABRAHALL observed a "slow change of the zero-point, 

 continuous throughout." 



It would appear that BAILEY and LAMB|| were more fortunate than some other 

 observers have been. They state that during their determinations of the atomic 

 weight of palladium, "the extreme variation of the zero-point did not exceed 0'3 

 division in two weeks." The sensibility (ll'l) of their balance was, however, 

 somewhat low. 



DEAN!! " found that the weighings could be made quite easily to a tenth of a 

 milligramme, provided that the final weighing was not made until half-an-hour after 



the tube and weights had been placed upon the pans of the balance That 



the half-hour interval needed was not on account of hygroscopic moisture absorbed 

 from the air by the material is shown by the fact that the weighings taken 

 immediately were always greater than those taken later." 



For the re-determination of the atomic weight of nitrogen, GRAY*"' used a balance 

 by SARTOBIUS. The instrument was placed in a cellar and screened by a tin box. 



* 'Phil. Trans.,' 1856, p. 765. 



t ' Mean Density of the Earth,' pp. 99 and 156. Griffin and Co., 1894. 

 J 'J. C. S.,' 1885, p. 116. j. C . S.,' 1892, p. 661. 



'J. C. S.,' 1892, p. 752. IT . j. C . S .,' 1900, p. 128. 



'J. C. S.,' 1905, p. 1608. 



** 



